anyone here bullish on state machine stuff? I want to be indoctrinated

me.

yeah here's a solid video:

peep Hillel Wayne's blog posts (eg https://www.hillelwayne.com/post/formally-specifying-uis/) and sketch.systems

Nice. I’ve been wondering – are state machines / state charts a category? They seem perhaps good for modelling session type like ideas too? And generally, all sorts of concurrent behaviour.

There's a whole community for that! https://spectrum.chat/statecharts/general?tab=posts

For clarification, I mean “category” in the mathematical sense, rather than “a subject / topic”! :-)

@Garth Goldwater I had just watched a different talk of his prior to making my post 😛 I'll look at this one too @Max Krieger That's where I first found out about them, actually. @stevekrouse Neat. Thanks. I will seek indoctrination

@Benjohn Barnes Not entirely sure if it's a category but category-friendly people work on Petri Nets https://johncarlosbaez.wordpress.com/2018/08/15/open-petri-nets-part-1/ https://blog.statebox.org/behavioral-programming-with-petri-nets-%C3%A0-la-functional-way-smart-contracts-96d801cef8cc

I wanted to add, in particular, why I'm surprised I don't see people on here talking about this more often. It seems to me that statecharts are showing completely essential complexity. Make a statechart for a GUI -- can it be reduced whatsoever? They're also intuitive to read for even non-programmers

@Benjohn Barnes Statecharts are hypergraphs with a single edge potentially leading on to multiple states. So its a generalization of a graph which I think can be partitioned to get a lattice of possible states. @Deklan Webster I have the same hunch too, I might report back in this early next year. Have a few systems to be built in this paradigm lined up. An open question I think is on how to do scheduling and other complex patterns on top of statecharts.

Some future of coding projects with statecharts: http://interstate.from.so/ http://web.media.mit.edu/~jacobsj/#db

@Deklan Webster Definitely agree they seem to “cut to the chase” of a lot of essential complexity allowing enormous removal of accidental complexity.

@stevekrouse Nice find on that dynamic brush tool. Do you know of any art tools/environments that use FRP paradigm in a similar manner (apart from Conal's work) ? I am constantly finding state machines to be the dual (duel?) of FRP paradigm in that they do localized reasoning of discrete events while FRP deals with global reasoning of continuous intervals.

There are many, many flavors of FRP, some of which are pull, some of which are push, and some of which as push-pull. Certain flavors of FRP is about global reasoning, but the original flavor was about local reasoning, and both continuous and discrete entities. Here's an example of a JS FRP library in the Conal style: https://github.com/funkia/turbine

@stevekrouse Knew I coupled something wrongly there. Would this stand up to scrutiny if I change the bottom axis to Transformative / Reactive? About the continuous vs. discrete dichotomy, I think continuity is a superset of discrete, so that subsumes considerations for discrete ones. And can't wrap my head around why FRP in the original version was about localized reasoing, which reminds me that I should read your FRP paper again.

I have actually been wondering how to relate FRP (or just RP) to statecharts...

I don't understand what the Transformative / Reactive would stand for

@stevekrouse I presume dataflow in pull (transform) or push (react)?

@stevekrouse Coopted that dichotomy from the original Statecharts paper by Harel:

Resurrecting this thread because I just remembered a paper my mentors wrote http://www.cs.cmu.edu/~aldrich/papers/onward2009-state.pdf